I'd hazard a guess that the first peak in the graph is from atmospheric testing of fusion weapons, and the second is from scientific, medical and commercial use of helium-3. I'm not sure why it would fall from 1988 to 1996, though.
Language Log discussed a certain behavior of the Google Ngram Viewer (which I assume is the source for this graph) a couple of years back. The entry is here.
http://languagelog.ldc.upenn.edu/nll/?p=2848
Before reading it, you might want to try doing a Ngram search for "funk", preferably from at least 1600 onwards. Can you guess what is going on with the results? (The article explains it.)
Well, I recognise it as a Google graph of the frequency of the phrase over time. I don't know much about that particular Google feature and have no idea what sort of corpus you're running it over - science fiction? all fiction? all books? all available printed and electronic text?
Probably not that last option, come to think of it, because the shape of the graph suggests that the absolute numbers involved are pretty small; I'd guess that the horizontal lines represent only about 5-10 in absolute numbers. So don't read too much into small random variations. I suspect that the graph from about 1973 to the present is statistically indistinguishable from a horizontal straight line at any reasonable significance level; ditto 1956-70.
As for the big drop in the 1970s, my guess would be that it reflects the general disillusionment about fusion power that started around that time.
Diving into the spike from 1959-1961, I suggest the surge may be due to some combination of cryogenic effects, the emergence of isotopic-abundance techniques, rising interest in nuclear fusion, and generally living in the Atomic Age.
The effects of Santarius, et al can clearly be seen circa 1990, although the general level is also propped up by interest in Big-Bang nucleosynthesis.
If you do "He3" you get a major spike in the 60s, a steady drop off towards the 80s, followed by a sudden smaller square blip for most of the 80s that then drops off before the 90s and a drop back down to 1940s era levels of use.
He3 also returns 3.2e-5 results vs. those 3.7e-6 results.
2012-03-29 06:28 am (UTC)
2012-03-29 06:53 am (UTC)
Another curious thing you might enjoy
(Anonymous)
2012-03-29 07:33 am (UTC)
http://languagelog.ldc.upenn.edu/nll/?p=2
Before reading it, you might want to try doing a Ngram search for "funk", preferably from at least 1600 onwards. Can you guess what is going on with the results? (The article explains it.)
2012-03-29 08:02 am (UTC)
2012-03-29 09:34 am (UTC)
2012-03-29 09:35 am (UTC)
2012-03-29 09:46 am (UTC)
Probably not that last option, come to think of it, because the shape of the graph suggests that the absolute numbers involved are pretty small; I'd guess that the horizontal lines represent only about 5-10 in absolute numbers. So don't read too much into small random variations. I suspect that the graph from about 1973 to the present is statistically indistinguishable from a horizontal straight line at any reasonable significance level; ditto 1956-70.
As for the big drop in the 1970s, my guess would be that it reflects the general disillusionment about fusion power that started around that time.
2012-03-29 12:04 pm (UTC)
2012-03-29 12:43 pm (UTC)
Setting the smoothing to zero:
Diving into the spike from 1959-1961, I suggest the surge may be due to some combination of cryogenic effects, the emergence of isotopic-abundance techniques, rising interest in nuclear fusion, and generally living in the Atomic Age.
The effects of Santarius, et al can clearly be seen circa 1990, although the general level is also propped up by interest in Big-Bang nucleosynthesis.
2012-03-29 02:41 pm (UTC)
He3 also returns 3.2e-5 results vs. those 3.7e-6 results.
2012-03-29 02:44 pm (UTC)
2012-03-29 01:06 pm (UTC)
2012-03-29 01:12 pm (UTC)
This Just In
2012-03-29 03:42 pm (UTC)
Re: This Just In
2012-03-29 03:50 pm (UTC)
Re: Re: This Just In
2012-03-29 04:06 pm (UTC)
(ahem)
Tralphium has been discovered!
2012-03-30 02:01 am (UTC)